The control of electronic devices has traditionally required the manual operation of dials, buttons, keyboards, keypads, and the like. However, such manual operation can be impractical, awkward, and inconvenient. For example, the terminal can be hooked to a belt underneath layers of clothing or a user might be wearing gloves, in which access to mechanical interfaces on the terminal is difficult and/or time consuming. Additionally, as technology progresses, manufacturers of electronic devices are continually reducing size of the devices, and thus, also reducing size of keyboards, keypads, or other mechanical interfaces. Such size reduction further hampers the ability of users to input accurate information.
Speech recognition technology has been developed for use in electronic devices in response to the aforementioned problems. In general, conventional speech recognition technology comprises a mechanism for receiving an input voice signal, comparing the input voice signal with stored voice signals, and determining if the input voice signal is sufficiently similar to any of the stored voice signals. If there is a match between the input and stored voice signals, instructions or other data is generated by the device.
However, a disadvantage with speech recognition is the need to “train” the speech recognition system to understand a specific vocabulary, language characteristics, and characteristics of the voice of the speaking person. The complexity increases with the size of the vocabulary that has to be recognized in any particular instance of the dialog with the application. Further, whether using speech recognition for command and control, having a telephone conversation, or performing dictation, there are instances in which speech commands are inadequate. For example, if a user is wearing a terminal and a speech recognition engine and decides to take a break, to maintain battery life, the system should suspend as many operations as possible. Yet, if the speech recognition system is suspended, which requires a significant amount of processing and power, the user has no method by which to re-activate the recognition, other than to remove the terminal from his belt and press buttons and/or keys.